In recent years, as video technologies develop, there are more and more video types, such as 2D videos, 3D videos and panoramic videos. With application of virtual reality devices, users can experience super large screen effect at home similar to that produced by a movie theatre, and panoramic videos even bring an immersed experience. Videos of different types need corresponding processing when being played. For example, playing 3D videos requires separate images for the left and right eyes and proportion adjustment; playing panoramic videos requires playing in 360 degrees. However, when devices play local 2D videos, 3D videos and panoramic videos, users do not know the video type, and can only select a type by themselves, which is inconvenient. Although some playing clients provide automatic identification functions for video types, there is a problem of inaccurate identification.